Please, update the name of the model to “veo-3.0-generate-preview” which means you must remove the -001 as per the latest documentation:
https://cloud.google.com/vertex-ai/generative-ai/docs/video/generate-videos-from-an-image
We’re currently seeing this error:
Hi @Joaquin_Carretero - welcome to the community!
Could you confirm whether you’re using the Gemini app or the Google Vertex AI app?
Both of those have access to the Veo3 model.
I am using the “Google Gemini Ai” node. Check out screenshot below
It is very time critical for us, please @DavidGurr_Make let me know if you could fix this. In short, the model name you are using for Gemini create videos is “veo-3.0-generate-preview-001”. This is wrong, you should make it “veo-3.0-generate-preview”. Until then, everyone in your platform is getting errors.
Hi Joaquin.
The Gemini app is managed by an external Solution Partner. I reported the issue to them and I’m expecting it to be resolved soon.
Either they or I will reply here once resolved.
In the meantime, I can confirm that the Google Vertex AI app uses the correct model name. You might be able to use this as a workaround if your need is urgent.
Hi @DavidGurr_Make, Hi @Joaquin_Carretero,
thank you for making me aware!
The change request has been submitted and should be live soon. Maybe @DavidGurr_Make can push this internally 
You can always click on “map” and adjust the model name if that’s needed (see screenshot). That’s probably the fastest solution
Best,
Richard
I can confirm that it’s live now!
Hi Joaquin.
Does the VEO 3.0 Preview model (not the fast one) work OK?
Can you confirm that you applied for access to Veo3 through the Google waitlist and have been granted access?
I see what you mean… It seems that i don’t have VEO3 access. Although, i work at Google. Please, let me know who’s a point of contact so i can contact directly for approval. Send me an email, please!
The only information we have is that you can apply via the waitlist … and then wait!
The latest information we have on this is that the Veo 3 is currently only available through the Google Vertex AI API, not the simpler Gemini API.
As a result, you may find that you can access it through the Google Vertex AI app in Make.
We haven’t yet been granted waitlist access, so we’re unable to check for sure. If that works for you. please let us know!
In fact … we’ve just noticed that the API waitlist form now says:
Veo 2 in Vertex AI text-to-video is now publicly available. This form is to request access to additional limited access for Veo 3 Image-to-Video.Veo3 Text-to-Video is now in Public Preview, so you can access it without any allowlisting requirements.
We’ve verified this and can confirm that Veo 3 Text-to-Video is now publicly available through the Google Vertex AI app in Make!
This does have a slightly more complex setup than using the Gemini app, requiring you to create a Google Cloud Project with the Google Vertex AI API enabled.
One further note … it seems that the Veo 3 model is only available for Google Cloud Projects using the us-central1 region.
Hello David, it seems to be working only via Vertex Inetrgation. Problem hgere is that the videos are only produced in 720p.
Is there a fix for the authentication problem with the build in veo3 Tool or any other fix to create Videos in 1080p
Thanks a lot!
Hi, and welcome to the Community!
Veo3 should be available in the Gemini AI app since last Monday (August 4). If you find that’s not working for you, please post details. But I’m pretty sure that Veo3 through the Gemini API (rather than Vertex API) is limited to 720p.
I’ll raise an enhancement request to add the resolution parameter to the Vertex AI Generate a Video module when veo3 is selected. I’ll report back here when I hear an update on release status.
